JPH02117289A - Variable length encoding device - Google Patents

Abstract

PURPOSE:To prevent the generation of overflow by shuffling a picture signal in a specified range, after that, blocking the picture signal and executing variable length encoding for the unit of the block so that a fixed length can be obtained. CONSTITUTION:The shuffling is executed in a shuffling part to the picture signal which is inputted from an input part 1. In a blocking part 3, the output of a shuffling part 2 is partitioned for each certain specified block and blocked. In a variable length encoding part 4, blocked data are completed in each block and the variable length encoding is executed so that the fixed length can be obtained. Then, the data are outputted to an output part 5. Thus, since the information quantity of a picture is dispersed by shuffling the data to be inputted, the information quantity to be included in each block is made almost equal. Accordingly, since the sum of a code length is almost equal in a case that the variable length encoding is executed for each block, data quantity control to fix the data quantity in each block can be easily executed as well.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a variable length encoding device used for highly efficient encoding of image information.

BACKGROUND OF THE INVENTION High-efficiency encoding technology has become important as image signals are digitized. In order to increase the efficiency of high-efficiency encoding, it is necessary to allocate more bits to areas with a large amount of information and fewer bits to areas where the amount of information is small. For this reason, variable length coding is often used. Table 1 is a correspondence table of variable length encoding for 3-bit data (0, 1, , , ?).

Table 1 In this example, 2 bits are used for 0 and 1, 3 bits are used for 2 and 3, and 4. 5. 6. 4 bits are assigned to 7. 0.10 occurrence probability is 4°5, 6. 7
If it is considerably larger than , the average number of bits after encoding will be less than 3 bits, and the effect of variable length encoding will appear.

In conventional variable length encoding, orthogonally transformed signals and prediction error signals are manually encoded in the order in which they are input. Therefore, to control the amount of data, variable-length encoded code words are stored in the buffer 7-, the amount of data in the buffer is detected, and the encoding is controlled so that the amount of data is constant within a specific range. Was. Further, as a control method, when the data in the buffer increases, the rounding at the time of encoding is increased, and when the amount of data decreases, the rounding is decreased.

Problems to be Solved by the Invention However, in variable length coding, if even a single bit error occurs, word synchronization is lost, and all information after that error is lost. Therefore, it is necessary to control the length to be constant within a specific small range (block) and reset error propagation within that range.

However, if the size of the blocks is reduced in a configuration like the conventional example, the dispersion of the amount of information contained in the image for each block becomes large. For this reason, blocks with a large amount of information will overflow the amount of data to be transmitted.
It becomes greatly distorted.

Conversely, in a block with a small amount of information, the amount of data to be transmitted is too small, so dummy data must be transmitted.

It is an object of the present invention to solve the problems of such conventional variable length encoding devices.

Means for Solving the Problems The present invention as claimed in claim 1 provides a blocking means for shuffling an image signal in a specific range and then forming blocks when variable-length encoding an image signal; and the blocking means. This is a variable length encoding device characterized by comprising variable length encoding means for variable length encoding the blocks obtained in the above so that each block has a constant length.

The present invention according to claim 2 provides a method for variable-length encoding a component image signal using a luminance signal and a color difference signal or R, G,
A blocking means for collectively forming blocks of B signals at a specific ratio, and a variable length encoding means for variable length encoding the blocks obtained by the blocking means so that each block has a constant length. A variable length encoding device is characterized by comprising:

Effect of the Invention With the above-described configuration, the present invention shuffles the image signal and then blocks it, so the information type in each block is distributed and becomes almost constant.

Therefore, even if variable length coding is performed using small blocks, the amount of information in each block is constant, so the above problem does not occur.

Further, in a normal image, the amount of information of the luminance signal and the amount of information of the color difference signal are not necessarily proportional. Therefore, by combining the luminance signal and the color difference signal into a block as described above, the amount of information of both signals complement each other, which becomes the TIT function, and even small blocks can be efficiently encoded.

Example The present invention will be explained in detail below.

FIG. 1 shows a first embodiment of a variable length encoding device of the present invention. In FIG. 1, 1 is the manual part of the present invention, 2 is the shuffling part, 3 is the blocking part, 4 is the variable length coding part,
5 is the output part of this device. In this device, a shuffling section 2 performs shuffling on an image signal input from a human-powered section 1. Shuffling refers to once storing manually input data in a RAM, and then randomizing it so that data that was continuous when manually input is output in a dispersed manner as much as possible when read. In the blocking section 3, the output of the shuffling section 2 is divided into certain intervals and divided into blocks. Then, in the variable length encoding section 4, the block data is variable length encoded so that each block is completed and has a constant length, and is outputted to the output section 5.

In the apparatus shown in FIG. 1, the amount of information in the image is distributed by manual data shuffling, so that the amount of information contained in each block is approximately equal. Therefore, since the sum of the code lengths when each block is variable-length encoded is approximately equal, data amount control to keep the data amount constant for each block is also facilitated. Therefore, overflow is less likely to occur.

FIG. 2 shows a second embodiment of the invention. In FIG. 2, 6 is the manual part of the present invention, 7 is the part that mixes the luminance signal and color difference signal at a fixed ratio, 8 is the blocking part, 9 is the variable length coding part, and 10 is the output part of this device. be. In this device, a luminance signal and a color difference signal are separated and manually input from a human input section 6. The luminance signal/chrominance signal mixing section 7 mixes the luminance signal and the chrominance signal at a certain ratio and outputs the mixed signal. The mixed data is divided into blocks by predetermined intervals, which are blocking portions 8 . Then, in the variable-length encoding part 9, variable-length encoding is performed so that the block data is completed in each block and has a constant length, and the output part 1
Output to 0.

In a normal image, the amount of information in the luminance signal and the information in the color difference signal are
are not necessarily proportional. For example, in a black and white image, even if the luminance signal has a large amount of information, the color difference signal has almost no amount of information. The opposite example is also possible. For this reason, as in the device shown in Fig. 2, by combining the luminance signal and color difference signal into a block, when the amount of information in the luminance signal is large and the amount of information in the color difference signal is small, it is possible to compensate for each other's information amount. Can be done. Of course, the same applies when the information surface of the luminance signal is small and the information amount of the color difference signal is large. Thereby, even small blocks can be encoded efficiently.

Also, this method uses not only luminance signals and color difference signals, but also R signals.

It can be applied to C and B signals as well.

Although the explanation has been made using two embodiments, it is also possible to use these two circuits simultaneously.

Effects of the Invention According to the present invention, with the above-described configuration, the image signal is shuffled and then divided into blocks, so that the amount of information in each block is distributed and becomes approximately constant.

Therefore, even if variable length coding is performed using small blocks, the amount of information in each block is constant, so problems such as overflow do not occur much.

Further, in a normal image, the amount of information of the luminance signal and the amount of information of the color difference signal are not necessarily proportional. Therefore, by combining the luminance signal and the color difference signal into a block as described above, it is possible to complement each other in their information amounts, and even small blocks can be efficiently encoded.

These allow variable length coding to be completed within a small range, thereby reducing the influence of error propagation even if an error occurs on the transmission path.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a first embodiment of the invention, and FIG. 2 is a block diagram of a second embodiment of the invention. 291. Shuffling part, 3. ,,blocking part,
4. , , variable length encoded part, 7. ,,luminance signal color difference signal mixing part

Claims (2)

[Claims] (1) When variable-length encoding an image signal, a blocking means that shuffles the image signal within a specific range and then blocks it, and a block unit that is constant for the blocks obtained by the blocking means. 1. A variable-length encoding device comprising: variable-length encoding means for performing variable-length encoding such that the length of the variable-length encoding device becomes long. (2) Blocking means for collectively forming blocks of luminance signals and color difference signals or R, G, and B signals at a specific ratio when variable-length encoding component image signals; 1. A variable-length encoding device comprising variable-length encoding means for variable-length encoding a block so that each block has a constant length.